Models and voices

Models

Case	Repository	Approx total download
.nanoInt8	KittenML/kitten-tts-nano-0.8-int8	28 MB
.nano	KittenML/kitten-tts-nano-0.8	59 MB
.micro	KittenML/kitten-tts-micro-0.8	44 MB
.mini	KittenML/kitten-tts-mini-0.8	83 MB

let config = KittenTTSConfig(model: .nanoInt8)
let tts = try await KittenTTS(config)

KittenModel exposes its Hugging Face repository, base URL, ONNX filename, voices.npz filename, approximate download bytes, display name, and voice speed prior.

Voices

Case	Raw ID	Character
.bella	expr-voice-2-f	Warm and expressive
.jasper	expr-voice-2-m	Clear and conversational
.luna	expr-voice-3-f	Calm and smooth
.bruno	expr-voice-3-m	Deep and steady
.rosie	expr-voice-4-f	Bright and friendly
.hugo	expr-voice-4-m	Authoritative
.kiki	expr-voice-5-f	Lively and energetic
.leo	expr-voice-5-m	Relaxed and natural

for voice in KittenVoice.allCases {
    print(voice.displayName, voice.id, voice.isFemale)
}

Nano models apply a model-specific voice speed prior; Micro and Mini return 1.0. The result's effectiveSpeed records the combined value.